View
222
Download
0
Category
Preview:
Citation preview
NASA/CR.--_"_,--.-- 207909
Final Report
1
i/# - /
Microgravity Science and_ Applications Division
Fluid Physics Program
The Circular Hydraulic Jump in Microgravity
NASA Grant NAG 3-1627
Principal Investigator:
Period:
C. Thomas Avedisian
Sibley School of Mechanical and Aerospace EngineeringCornell University193 Grumman HallIthaca, New York 14853-7501
phone: 607-255-5105FAX: 607-255-1222email: cta2 @ cornell.edu
June 24, 1994 to June 23, 1996
Cost:
https://ntrs.nasa.gov/search.jsp?R=19980048927 2018-04-26T21:53:11+00:00Z
Abstract
This report summarizes the key experimental results and observations that were obtained
under NASA grant NAG 3-1627 from the Fluid Physics Program. The Principle Investigator was
Thomas Avedisian. In addition a half-time post-doctoral associate, Ziqun Zhao, was funded for
half year. The project monitor was David Chao of the NASA-Lewis Research Center in Cleveland,
Ohio. The grant period was originally for one year at $34K and a no-cost extension was applied for
and granted for an additional year.
The research consisted of an experimental study of the circular hydraulic jump (CHJ) in
microgravity using water as the working fluid. The evolution of the CHJ radius was measured
during a sudden transition from normal to microgravity in a drop tower. The downstream height of
the CHJ was controlled by submerging the target plate in a tank filled with water to the desired
depth, and the measurements are compared with an existing theory for the location of the CHJ.
Results show that the CHJ diameter is larger in microgravity than normal gravity. The
adjustment of the CHJ diameter to a sudden change in gravity occurs over a period of about 200ms
for the conditions of the present study, and remains constant thereafter for most of the flow
conditions examined. For flow conditions that a CHJ was not first established at normal gravity but
which later appeared during the transition ti5 microgravity, the CHJ diameter was not constant
during the period of microgravity but continually changed. Good agreement between measured and
predicted CHJ radii is found for normal gravity CHJ radii, but comparatively poorer agreement is
observed for the CHJ radii measurements in microgravity.
1. Objectives
The purpose of this project was to study liquid jet impingement in microgravity. The
emphasis was on the hydraulic jump phenomenon. The objectives were the following:
1) record the evolution of a circular hydraulic jump (CHJ) upon a transition from Earth
normal gravity (G= 1) to microgravity, G<< 1-t;
2) make quantitative measurements of the CHJ diameter in microgravity;
3) measure the curvature of the free liquid surface across the CHJ during the transition
from normal to microgravity; and
4) compare the CHJ diameter measurements with published predictions.
Figure 1 is a schematic diagram that shows the basic features of an impinging liquid jet that undergoes
a hydraulic jump. The parameters were the liquid flow rate, jet orifice diameter, and the downstream
liquid height.
The decision to focus on the transition from G=I to G<<I was made because of the available
experimental run time of just over ls as discussed in Section 3. The transition of the CHJ from
normal to microgravity was considered to be the easiest way to document gravity's effect on the CHJ
without pursuing a more extensive study of liquid jet impingement in which the jet would first be
established in microgravity.
2. Importance of the Circular Hydraulic Jump
Impingement of a circular liquid jet onto a surface is important in a variety of processes:
impingement cooling of electronic devices, materials in manufacturing processes, laser mirrors, and
aircraft generator coils; and vapor absorption refrigeration cycles. A feature of impinging liquid jets
is their potential for dissipating high heat transfer rates. For example, the highest steady state heat
fluxes reported in any configuration were achieved by a water jet impinging onto a plasma heated
surface (Liu and Lienhard 1993). A concern for jet impingement is the potential for the formation of
a circular hydraulic jump downstream of which the liquid velocity, and thus heat transfer, are
tG=g/g o where go -9.8m/s2- In the remaining text, the designation 'G=I' means Earth's normal gravity and 'G<<I'
means 'microgravity' which for the proposed experiments will mean any G value in the range 10-4<G<I0 -2 based on
the drop tower package used (Avedisian et al. 1988)
4
reduced. It is, therefore, important to predict the location of the CHJ (the jump diameter, Dh) and
the parameters upon which it depends to avoid the reduction of heat transfer. Gravity figures
prominently in both the location of a CHJ and the curvature of the free liquid surface across the
jump. The CHJ occurs where the expanding liquid jet undergoes a transition from a 'supercritical'
to a 'subcritical' flow, in the sense of a suitably defined Froude number being greater or less than
unity, respectively.
Existing formulations (see Section 5.2) for predicting Dh (Middleman 1995) show that as
gravity is reduced Dh should increase. As a result, heat transfer rates may be extended in
microgravity to larger diameter. Our results confirm this general trend, though quantitative
agreement is not demonstrated. The flow downstream of a CHJ in microgravity is significantly
different from the downstream flow at G=I. Capillary waves form in the downstream flow in G<<I
and a more gradual transition across the jump occurs from the toe of the jump (see fig. 1) to the
downstream flow. The origin of these waves is speculated to be due in part to the rapid increase in
curvature created by the sudden transition from G= 1 to G<<I which we demonstrate in this report,
or, as speculated by Hornung et al. (1995), to vorticity generation at the CHJ.
This proposal was envisioned to be the first part of a larger effort on liquid jet impingement
in microgravity that would have included fluid mechanics and heat transfer. In the first part,
described in this report, experiments to characterize the fluid mechanics of an impinging jet in
microgravity were carried out. The primary focus was on measuring the evolution of Dh during the
transition from G=I to G<<I: the surface curvature across the CHJ, film thickness across the CHJ,
and Dh were measured. The second (future) part was to continue these isothermal measurements
and to pursue a study of heat transfer to an impinging liquid jet in microgravity.
3. Brief Review of Prior Work
An extensive literature exists on jet impingement in general, and the hydraulic jump in
particular. Standard fluid mechanic textbooks provide basic formulations for planar inviscid
hydraulic jumps (e.g., Allen and Ditsworth 1972; Fox and McDonald 1992). For the CHJ,
Nirapathdongporn (1968) reviews the literature prior to 1968. Errico (1986) reviews some of the
latertheoriesandpresentsmeasurementsof theCHJdiameteratG=1whichshowvariousdegrees
of agreementwith analyses.Theroleof surfacetensiononstabilizationof theCHJjump surfaceat
G=I wasdiscussedby Liu andLienhard (1993), who showedthat increasingthe fluid height
downstreamof thejump increasedtheeffectof surfacetensionat thefreeliquid surface.
A suitablydefinedFroudenumber,Fr = U/c where'c' is theappropriatewavespeed(e.g.,
c=(gh)1/2from 'shallow water' theory (Milne-Thomson 1979,p. 443) is an indicator for the
expectationof ahydraulicjump. Aboveacritical value,Frc,in thesenseof themeanliquid velocity
exceedingthewavespeed,a CHJwill alwaysoccurat someradial location in the expandingjet.
While Frc=l, a non-unity critical Froudenumberhasbeenargued(Rahmanet al. 1991b)to be
dependenton thevelocityprofileusedfor theanalysisof theCHJ.
A numberof numericalandanalyticalstudieshavepredictedthebehaviorof theCHJ at
G=0. Laminar (Rahmanet al. 1990, 1991a; Thomas et al. 1990) and turbulent (Rahman et al.
1991) jets have been analyzed and predicted to be free of a hydraulic jump at G=0. In the vicinity
of a jump at G=I, back-flow, separation, and eddies are observed in experimental studies and
predicted in early treatments of the problem (e.g., Craik et al. 1981; Tani 1948). No analogous
observations have been reported for G<<I. A 'classic' analysis by Watson (1964) results in a
closed-form solution to the laminar momentum and continuity equations for the CHJ radius where
an inviscid approximation is made for the stagnation region and the flow downstream of the jump,
the flow is self-similar upstream of the CHJ, and the limit of an infinite Froude number is assumed.
Some failures of the model are analyzed by Bowles and Smith (1992) who extend the analysis to
finite Fr and predict the film thickness shape near the jump. Numerical solutions by Chaudhry
(1994) accounts for the effect of turbulence and heat transfer and provide predictions of both the
CHJ diameter and film thickness downstream of the jump.
One experimental study is known on the problem of jet impingement at G<<I (Labus
1976). The focus was on studying the shapes assumed by a free liquid surface rolling off of the
edge of a plate on which a liquid impinged. A drop tower was used to create the microgravity
environment. A hint of what to expect for the CHJ diameter at G<< 1 was noted by the observation
6
of Labus(1976) thatno CHJ occurredduringany of the G<<I tests while they were a common
occurrence at G=I for nominally the same flow conditions. Reducing gravity thus appears to push
the CHJ off the target plate for the flow conditions examined.
4. Experimental Design
A microgravity environment for the present experiments was created by using a drop tower.
Central to the success was the ability to observe all desired features of the flow within the available
experimental run time. It was anticipated that the reaction time of the CHJ radius to sudden changes
in G would be much less than the available experimental time (on the order of ls for the facility
used). While this was verified for most of the conditions analyzed, for some it was not as
discussed in the next section.
The basic CHJ visualization set-up is shown in fig. 2. It consists of a flow-loop with the
following components: recirculating pump; plenum containing beads and a flow straightener (to
dampen the inlet flow - see fig. 3) and a removable orifice plate attached at one end; plexiglas
chamber within which the target plate is mounted; flow control valve; and cameras and lighting.
The height of the liquid level downstream of the jump is controlled by partially submerging the
target plate by raising the water level to the desired height above the plate. The experimental
conditions examined were the following. The downstream fluid height, hoo, was 2.0 mm, 4.0 mm,
6.0 mm, 10.0 mm, or 15.0 mm. The liquid flow rates were 2.39ml/s, 6.32ml/s, 9.93ml/s, or
26.47ml/s. To create laminar jets, a sharp-edged orifice was used. Orifice diameters of 1.22mm,
2.56mm and 3.83mm were used. They were machined into a stainless steel plate that could be
bolted to the bottom of the plenum (see fig. 3). The plenum was mounted such that the orifice was
7.62cm above the target surface. The diameter of the liquid jet, Dj, at a reference 2Dj above the
target surface was measured directly.
The target plate was a 6.35 mm thick and 23 cm diameter pyrex glass disk. It was painted
black on its back to reduce light reflection. The glass disk was attached to a plexiglass support disk
1.9 cm thick and 23 cm I.D. and mounted to one end of a circular plexiglass cylinder that was then
bolted to the containment vessel and sealed with o-rings. A mirror tilted to 45 ° under the circulation
tank(seefig. 2) allowedfor observationsof theundersideof thejet. This view wasaprimaryone
to measuretheCHJradius. Additional viewing angleswerealongtheplanethroughthe liquid to
showthecross-sectionalprofile shapeof theliquid film acrosstheCHJ,andanangledtopview for
globalfeaturesof theCHJ.Theworkingfluid waswateratroomtemperaturein all theexperiments.
To keepthe designsimple, theprimary meansof dataacquisitionwasphotographic. A
35mmNikon F3 camerawith MD-4 motordrive (operatedat 5 frames/sor 200msintervals)and
attached105mm NIKKOR macrolenswasusedto recordimagesat all threecamerapositions.
Approximatelyfive 35mmexposureswereobtainedin theperiodof G<<I. Video imagesusinga
COHU CCD camera(30 images/s)with attached28mmNIKKOR macrolenswereusedto record
theevolutionof theCHJ in thetransitionto microgravity,and30videoimageswereobtained.All
camerasandlensesweresecurelymountedto preventdamageby theshockof theimpact.Lighting
wasprovided by halogenlamps. The CHJ diameterwasmeasuredfrom the video imagesof
underneathviews usinga 'videocaliper'(Video Caliper306,ColoradoVideo Inc.) placedon the
videomonitorthatwascalibratedWithaprecisionruler(SchaedlerQuinzel,Inc.,USA)addedto the
image. Sideviews (from the 35mmcamera)wereusedto obtainthecross-sectionalshapeof the
liquid film acrossthe CHJ. The sideview imageswere fed into a MAC-baseddataacquisition
system(AUTOMATIX ImageAnalysisProgram)andanoperator-selectedgrayscalewasusedto
identify thevariousboundariesinvolved.A 19.05mm diameterball bearingimageconvertedthe
sideview pixel countto lengthwith aprecisionof about+66 lam.
The experimental procedure was as follows. Flow conditions were first set so that a CHJ
existed at G=I (several conditions were also examined in which a CHJ did not exist at G=I but
which later appeared during the transition to G<< 1). The instrumentation package was then released
into free fall by deactivating the magnet which initially held the package to the support ceiling. With
the various cameras in operation, features of the impinging jet were then recorded during the flight
of the package. Analysis of the photographic images was done as per above.
5. Discussion of Results
5.1 Visualization
Fig. 5 shows two representative photographs of a CHJ. The upper photograph is for a jet at
G=I and the lower photograph shows the same jet 400ms into the period of G<<I. The
downstream fluid depth is 4 mm and the flow rate is 9.32ml/s. As is typical of jet impingement, the
upstream flow is a thin film which transitions to the downstream fluid depth over a short distance at
the CHJ. At G<<I, the downstream flow shows a series of concentric capillary waves that do not
exist at G=I. The transition of the flow across the jump is more gradual at G<<I than at G=I and
the liquid boundary at the CHJ appears to be pushed outward.
Fig. 6 shows a sequence of photographs of the change of the free surface across the jump at
five different times (spaced by 200ms) during the period just prior to microgravity (G=I) and into
microgravity (G<<I). The vertical line is the jet. The change of curvature across the jump is
clearly shown in fig. 6. After entering microgravity, the jump moves outward from the stagnation
point and the jump curvature increases. The source of the downstream waves in G<<I is due to the
abrupt increase of radius of curvature when gravity transitions from G=I to G<<I. It remains to
show that a flow downstream of a CHJ in microgravity would not exhibit this behavior if the CHJ
were first established in microgravity. This question will require a longer experimental flight time.
Fig. 7 shows bottom view photographs depicting the increase of the CHJ diameter in
microgravity relative to normal gravity. In these photographs the jet is directed outward toward the
plane of the viewer. The circular capillary waves are also shown in fig. 7.
The effect of downstream fluid height, hoo, on the CHJ is shown in the series of
photographs in fig. 8. For hoo=2 mm and Q = 6.32ml/s, a smooth jump with minimal downstream
disturbance is seen. During the transition to G<<I an abrupt change of curvature occurs at the toe
of the jump which forms a large wave in the downstream flow that appears like a 'hump'. This
effect is also shown in fig. 9 which is in the sequence of fig. 5 taken 200ms into G<<I. At h,,o=4
mm and Q = 6.32ml/s (fig. 8) the increase in CHJ diameter is also evident, as is the thinning of the
downstream flow due to the increase in curvature.
At lower flow rates, the CHJ moves closer to the stagnation point. This effect is shown in
fig. 10. For hoo = 2 mm (fig. 10) the capillary waves exist even at normal gravity (compare with
8
9
fig. 8). At 4 mm and 2.39ml (fig. 8) the jump is very close to the stagnation point but the surface
of the downstream flow appears very turbulent. Since in G<<I the surface curvature across the
jump increases - the transition across the jump is more gradual - the primary fluid effect of G
appears to be in the hydrostatic pressure of the downstream flow. This point is discussed further
in the next section.
5.2 Quantitative Measurements of the CHJ Diameter.
Measurements of the evolution of the CHJ diameter were made from the video images using
a computer-based analysis system. Figs. 11 shows the effect of flow rate on the CHJ at two
different flow rates for the 2.56 mm diameter orifice and hoo=4 mm. At t<0s, G=I and a steady
CHJ exists. At t=0s, the instrumentation package is released into free fall and the period of
microgravity begins. As shown from the measurements in fig. 11, enough time appears to exist for
the CHJ to reach a steady position at G<<I.
The effect of jet diameter on the CHJ is shown in fig. 12. Increasing the jet diameter pushes
the CHJ outward. The adjustment of the CHJ to microgravity is clearly shown and again occurs on
a time shorter than the available run time of about 1.2s.
The effect of downstream fluid height on the CHJ diameter is shown in fig. 13 at both
normal and microgravity. The lines shown in the figure are placed to suggest trends. At normal
gravity, Dh decreases as ha increases. This effect is speculated to be due to the increase of
hydrostatic pressure in the downstream flow as hoo increases which would tend to increase the
inwardly directed radial pressure on the downstream side of the jump. The jump moves inward
where the fluid velocity will be higher to re-establish a balance of momentum across the jump. For
G<<I, however, there is no clear trend of CHJ diameter with hoo. If the hydrostatic pressure of the
downstream flow is predicted to effect Dh but measured trends with hoo are not predicted for G<< 1,
it may be that wall shear in the downstream flow which is neglected in the analysis is in fact
important in G<< 1. For G=0 Dh should be infinite so the small but non zero gravity level in the
present experiments may unmask the role of friction in the downstream flow on Dh.
While this investigation was focused on experiment, the plethora of analyses of the
hydraulic jump motivated applying someof theseanalysesto the data obtainedduring this
investigation. Most of the publishedanalysishasbeenassociatedwith 'planar'jumps. For the
CHJ,comparativelylittle workhasbeenreported.Middleman(1995)reviewssomeof thetheories.
A simplifiedviewpointof theoriginof aCHJ isof aflow slowedby friction atthewall (and
theincreasingflow areadueto theradialgeometry)until apointis reachedatwhichtheflow cannot
adjustto the changingdownstreamconditionswithout experiencinga 'jump' in film thicknessto
satisfy massconservation. In anothertheory, the liquid jet is assumedto expandunder the
influenceof anadversegravitationalpressuregradientandwill eventuallyseparate(Tani 1948)asa
result of a nonlinearinteractionbetweenthe wall shearstress,surfacetensionand thepressure
gradientacrossthefilm (BowlesandSmith 1992;GaudierandSmith 1983).This interactionleads
to an upstreaminfluence and the CHJ is the result that allows an adjustmentof the flow to
downstreamconditions.
The earliest, and most often quoted,analysesof a CHJ assumethat the upstreamand
downstreamflows areinviscid. Applying a mass and momentum balance across a control volume
that includes the CHJ (see fig.), and a force balance across the free surface of the jump, results in a
relation between Dh, curvature of the free surface, _c, (=I/R where R is the radius of curvature of
the fluid across the jump) and flow parameters as
nD h) _,h h- =2 g°(h_-
10
and
R
pGgoh_2
where 'R' is the radius of curvature and _ is the curvature. From eq. 1 as G decreases, Dh
increases; and, for given G, Dho_ 1/hoo 2. From eq. 2, _: _ G. The effect of G on Dh and _ is
qualitativelyverifiedby thedatashownin figs 11,12,and 14in whichDhis largerin microgravity
thannormalgravity,andthecurvatureacrossthejump is lessin microgravitythannormalgravity.
At G=0,Dhis infinite byeq. 1andthusshouldnotbeobserved,but this limit is animpracticalone
becauseof the impossibilityof experimentallycreatingpreciselytheG=0condition.
SinceG changesabruptly in thepresentexperiments,the suddendecreasein _ canbe the
sourceof wavesthat are clearly evident in fig. 5, 8, and 10 for hoo=2mmand t=O.2sin the
downstreamflow in microgravity.
Thequalitativetrendof how Dhdependsonh,,ofor givenG that ispredictedby eq. 1 is not
confirmedby the G<<I measurementsreportedhere. Fig. 13showsthat Dhexhibits at mosta
weakvariationwith hooashooincreasesin G<<I while eq. 1showsthat Dh shouldasymptoteto
infinity asG is reduced.This limit G is not followed by themeasurementsandthe discrepancy
couldbedueto neglectingfriction.
An extension of eq. 1 to include friction in the upstream fluid was first reported by Watson
(1964) who uses a boundary layer model to predict the velocity profile across the upstream film.
The results show that the flow becomes self-similar at some radial position away from the
stagnation point. This theory has the essential features of more advanced treatments (e.g., Bowles
and Smith 1992; Higuera 1994) and is applied to the present data. The result can be expressed in
non-dimensional form by introducing the parameters (Middelman 1995) Y -= RH/Froo+I/(2RH),
R-Dh/DjRe-I/3, the Reynolds number Re- 4Q/(rcDjv), H-2h,,JDj Re 1/3 and Froo-Uj2/(Ggohoo):
11
Y =.26/(R3+.287)
where the approximation hoo>>h has been made. The viscous limit is R---)0 in eq. 3. Y and R are
known from the experimental results: Dh as discussed previously, and hoo and Q as input
parameters (Uj=Q/[r_Dj2/4]). A reference location of 2Dj above the target surface was used to
measure Dj. For the water kinematic viscosity at 300°K we took v--8.95x10-7m2/s (Keenan et al.
1969). An average G of 1.1xl0 -2 is used over the free-fall distance for the unshielded falling
package.Thougheq.3 is notexplicit in thejump radius,it canbeusedto determinetheextentof
agreementbetweenmeasuredandpredictedCHJradii.
Fig. 15showsthevariation of Y with R. The inviscid limit (dotted line) clearly does not
predict our measurements while all of the G= 1 data are predicted reasonably well when viscosity
effects are included (eq. 2). On the other hand all of the measurements of Dh for G<< 1 are not well
predicted by eq. 3. These observations suggest that a physical process which is masked at G=I
may become dominant at G<< 1. Candidate processes include flow separation downstream of the
jump and the nonuniform downstream velocity that is not included in the analysis, or neglect of
viscous drag at the surface of the plate.
6. Conclusions
The major conclusions from this nominal one-year preliminary study are the following:
1) a steady CHJ can be established in microgravity;
2) the CHJ in microgravity is larger than at normal gravity, other flow conditions being
the same;
3) the fluid response time to re-establish a CHJ at G<<I from a jump at G=I is on the
order of 200ms;
4) the transition of the fluid film thickness across the CHJ boundary is more gradual in
microgravity than at normal gravity due to the reduction in hydrostatic pressure in the downstream
film;
5) the G=I measurements are well correlated by an existing formulation but the
microgravity measurements are not well predicted;
6) the CHJ diameter shows no clear trend with hoo in a microgravity environment; and
7) All G<<I downstream flow patterns showed capillary ripples.
12
Publications/Presentations:
The Circular Hydraulic Jump in Microgravity, Journal of Fluid Mechanics, submitted
13
7. References
Allen, T. and Ditsworth, R.L. (1972) Fluid Mechanics, pp. 291-291, McGraw-Hill, New York.
Avedisian, C.T., Yang, J.C and Wang, C.H. (1988) Proc. R. Soc. Lond A420, 183.
Bowles, R.I. and Smith, F.T. (1992) J. Fluid Mechanics. 242, 145.
Chaudhury, Z.H. (1964) J. Fluid Mechanics. 20, 501.
Craik, A.D.D., Latham, R.C., Fawkes, M.J. and Gribbon, P.W.F. (1981) J. Fluid Mechanics
112, 347.
Errico, M. (1986) "A Study of the Interaction of Liquid Jets with Solid Surfaces," Ph.D. Thesis,
University of California, San Diego.
Fox, R.W. and McDonald, A.T. (1992) Introduction to Fluid Mechanics, 4th edition, pp. 525-530,
John Wiley, New York.
Gajjar, J.S.B. and Smith, F.T. (1983) Mathematika 30, 77.
Higuera, F.J. (1994) J. Fluid Mechanics. 274, 69.
Jackson, G.S. and Avedisian, C.T. (1994)Proc. R. Soc. Lond., A446, 257-278.
Keenan, J.H., Keys, F.G., Hill, P.G., and Moore, J.G. (1969) Steam Tables, p. 114, John
Wiley, New York.
Labus, T.L. (1976) "Liquid Jet Impingement Normal to a Disk in Zero Gravity," Ph.D. Thesis,
University of Toledo.
Liu, X. and Lienhard, J.H. (1993) Exp. Fluids, 15, 108.
Middelman S. (1995) Modeling Axisymmetric Flows, Chapter 5, New York, Academic Press.
Nirapathdongporn, S. (1968) "Circular Hydraulic Jump", M.Eng. Thesis, Asian Institute of
Technology, Bangkok, Thailand.
Thomas, S., Hankey, W.L., Faghri, A., and Rahman, M.M. (1990) J. Heat Transf 112, 728.
Tani, I. (1948) J. Phys. Soc. Japan 4, 212.
Watson, E.J. (1964) J. Fluid Mechanics 20, 481.
D O
gj
liquid jet
hydraulic jumpcontrol volume
h(r)U2
stagnation_ _"
region _
Figure 1
r o
r
rh
steel aluminumtop
plate
105mmlens plexiglasssidewalls ..... ==
lens support
35 mm /_
camera body hoo
plenum
water level,
electromagnet
steel plate
hydraulic jump
halogen lamp
electronic flowmeter
28 mm aluminum
supportbracket
mirror
Figure 2
drain valve
flow controlvalve
--- aluminum
plate
waterinlet
flangeweld
lassbeads(3.5dia.)
152.40
meshscreen
-- PVCsleeve
O-rin< orificeplate(seebelow)
Plenum Design[numbers in mm; not to scale]
Figure 3
6-32 clearance
. (6 reqd.)
J
DO (= 1.22, 2.56 ,3.83, 5.08) O-ring groove
j.il L_"-'
57.15
! !
63.50
Detail of orifice plate[numbers in mm; not to scale]
Figure 4
G=I; h_o=4mm; Q=9.32ml/s
(a)
G<<I
(b)figure 5
!I
Figure 6
D o = 2.56 mm Q = 9.932 ml/s h_ = 4 mm(View from underneath traget plate)
t=0G=I
k
t = 0. 2 secG<<I
40 mmI I
t = 0.4 secG<<I
waves
figure 7
D O = 1.22 mm
hoo = 2 mm
Q = 6.32 ml/s
hoo = 4 mm
t=OG=I
t=0G=I
t = 0.2 secG<<I
t = 0.2 secG<<I
t = 0.4 secG<<I
t = 0.4 secG<<I
I I40 mm
figure 8
t = 0.6 secG<<I
Photograph of a CHJ taken 200ms after the period of microgravity. The flowconditions are identical to figure 5. The 'hump' just downstream of the jump is
caused by the rapid increase of curvature from G=I to G<<I.
Figure 9
hoo
D o = 1.22 mm
=2mm
Q = 2.39 ml/s
hoo = 4 mm
t=0G=I
40 mm
t=0G=I
t = 0.2 secG<<I
t = 0.2 secG<<I
t = 0.4 secG<<I
t = 0.4 secG<<I
figure 10
100
80
6O
40Oq
20
D = 2.56 mmo
' ' ' ' ' ' I _ ' _ I ' ' ' I ' ' ' I ' ' ' I ' ' '
G=I _ G<<I
-00-0-_0000000000-00-0
0 o
---0000 ....
transienl period
Q = 26.47 mils
• Q = 9.93 ml/s
••00•00000000-•0/
0 L I i I i _ i I i , I I l i L I i i , I , , * I J l l
-0.2 0 0.2 0.4 0.6 0.8 1 i.2
Time (sec)
Figure 11
80
70
6O
50
E
40e..
30
20
10
0-0.2
OOO O
h =4 mm; U.=3.1 _sj
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
D =3.83mmo
D =2.56 mmo
O., O _1, 41. _1' _, 41. 41, # O. • O _. O # _. _' _1, # • 41, _1, • 41, • O _ O _
41, #
##,### D =1.22 mmo
i i i I i i i I _ r i I i i i I i i i I i t , I _ i i
0 0.2 0.4 0.6 0.8 1
Time (sec)Figure 12
.2
e-
60
5O
4O
3O
20
10
00
Q=6.32 ml/s
........ G<<I
+ G=I
i no jump observed at G=I
t I I [ I ' J J t I B r l I I ] I I a ] i r r I j i i I t I i
2 4 6 8 10 12 14
h(mm)
Figure 13
16
3.5
3
2.5
_ 2E 1.5>- 1
0.5
0
-0.5
3.5
3
2.5
2
E 1.5
>- 1
0.S
0
-0.5
3.5
3
2.5
2E
E 1.5>- 1
0.5
0
-0.5
3.5
3
2.5
E 1.5
>- 1
0.5
0
-0.5
3.5
3
2.5
1.5
0.5
0
-0.5
3.5
3
2.5
P 1.5
>" 1
0.5
0
-O.S
_ .......... __qi__.___- .O.j _ _---_'_- Time ffi 0 ms
i.._.Z .................................................................................JS '1'0 ' 1; 20 25
- Time ffi 200 ms -
..........................................................................................................'lb .... l's '2b .... 2s
Time = 400 ms
_.,,,. =. ,.r_lP.dB_ i
_..- d.,,-._,=..._....-.e_. ..........................................................................................................=
s 1'o .... l's '2'o 2s
__:__ Time = 600 ms -
- ____-
- o.o...,"'__'__
_W OII am O
: :
s 1'o 1; ' ' zb ' ' zs
- Time ffi 800 ms -
,_._°o o.,O''.__ =d==o=* _=a= ow =" -:
oeeod pr
. . O "_Ip" 8°_DqD_Nllu°.-
..... J.gl .....................................................................
: is 1'0 l'S ' ' ' z'o zs
Time = 1000 ms
.==_ oo o o
s ' lb l'S .... 2'0 2sX (mm)
figure 14
10
1
0.1
0.01
0.001
0.00010.1
inviscid limit
i
i
:[]
[] Bg; jump at lg
Bg; no jump at Ig _ : '
\• lg
..... Watson (1964): r<ro
-- Watson (1964): r>ro
i I h i i i i
1
R
Figure 15
i i i L l L L L
I0
Recommended